{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 2,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7529411764705882,
      "grad_norm": 32.96015025657947,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.586988091468811,
      "memory(GiB)": 58.9,
      "step": 1,
      "token_acc": 0.6259816346033988,
      "train_speed(iter/s)": 0.005083
    },
    {
      "epoch": 1.0,
      "grad_norm": 32.96015025657947,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.6049340963363647,
      "memory(GiB)": 61.17,
      "step": 2,
      "token_acc": 0.6236321303841676,
      "train_speed(iter/s)": 0.007742
    },
    {
      "epoch": 1.7529411764705882,
      "grad_norm": 55.64718634987283,
      "learning_rate": 1.5e-06,
      "loss": 1.5901291370391846,
      "memory(GiB)": 64.93,
      "step": 3,
      "token_acc": 0.6228877679697352,
      "train_speed(iter/s)": 0.006317
    },
    {
      "epoch": 2.0,
      "grad_norm": 55.64718634987283,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.5883402824401855,
      "memory(GiB)": 64.93,
      "step": 4,
      "token_acc": 0.6282280894332428,
      "train_speed(iter/s)": 0.007441
    },
    {
      "epoch": 2.7529411764705882,
      "grad_norm": 27.240683196193338,
      "learning_rate": 2.5e-06,
      "loss": 1.584397554397583,
      "memory(GiB)": 64.93,
      "step": 5,
      "token_acc": 0.6237093456611612,
      "train_speed(iter/s)": 0.00659
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.240683196193338,
      "learning_rate": 3e-06,
      "loss": 1.575513243675232,
      "memory(GiB)": 64.93,
      "step": 6,
      "token_acc": 0.6160881463344375,
      "train_speed(iter/s)": 0.007327
    },
    {
      "epoch": 3.7529411764705882,
      "grad_norm": 29.981942202913814,
      "learning_rate": 3.5e-06,
      "loss": 1.5442345142364502,
      "memory(GiB)": 64.93,
      "step": 7,
      "token_acc": 0.6285074626865672,
      "train_speed(iter/s)": 0.006735
    },
    {
      "epoch": 4.0,
      "grad_norm": 16.74358565099932,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.4945019483566284,
      "memory(GiB)": 64.93,
      "step": 8,
      "token_acc": 0.6307132263718485,
      "train_speed(iter/s)": 0.007274
    },
    {
      "epoch": 4.752941176470588,
      "grad_norm": 20.913266515990472,
      "learning_rate": 4.5e-06,
      "loss": 1.456627607345581,
      "memory(GiB)": 64.93,
      "step": 9,
      "token_acc": 0.6366711923256694,
      "train_speed(iter/s)": 0.006824
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.913266515990472,
      "learning_rate": 5e-06,
      "loss": 1.4757978916168213,
      "memory(GiB)": 64.93,
      "step": 10,
      "token_acc": 0.6461680273900448,
      "train_speed(iter/s)": 0.007242
    },
    {
      "epoch": 5.752941176470588,
      "grad_norm": 16.04621922221096,
      "learning_rate": 4.99847706754774e-06,
      "loss": 1.350942611694336,
      "memory(GiB)": 64.93,
      "step": 11,
      "token_acc": 0.6594030608711711,
      "train_speed(iter/s)": 0.006886
    },
    {
      "epoch": 6.0,
      "grad_norm": 16.04621922221096,
      "learning_rate": 4.993910125649561e-06,
      "loss": 1.3158870935440063,
      "memory(GiB)": 64.93,
      "step": 12,
      "token_acc": 0.6653147813065452,
      "train_speed(iter/s)": 0.007229
    },
    {
      "epoch": 6.752941176470588,
      "grad_norm": 14.678773602579698,
      "learning_rate": 4.986304738420684e-06,
      "loss": 1.2742345333099365,
      "memory(GiB)": 64.93,
      "step": 13,
      "token_acc": 0.6654476469109277,
      "train_speed(iter/s)": 0.006922
    },
    {
      "epoch": 7.0,
      "grad_norm": 8.92125152971249,
      "learning_rate": 4.975670171853926e-06,
      "loss": 1.2416595220565796,
      "memory(GiB)": 64.93,
      "step": 14,
      "token_acc": 0.6752563337492541,
      "train_speed(iter/s)": 0.007215
    },
    {
      "epoch": 7.752941176470588,
      "grad_norm": 14.395749147533214,
      "learning_rate": 4.962019382530521e-06,
      "loss": 1.1947765350341797,
      "memory(GiB)": 64.93,
      "step": 15,
      "token_acc": 0.6894364179593173,
      "train_speed(iter/s)": 0.006955
    },
    {
      "epoch": 8.0,
      "grad_norm": 14.395749147533214,
      "learning_rate": 4.9453690018345144e-06,
      "loss": 1.1081597805023193,
      "memory(GiB)": 64.93,
      "step": 16,
      "token_acc": 0.6838079628491429,
      "train_speed(iter/s)": 0.007209
    },
    {
      "epoch": 8.75294117647059,
      "grad_norm": 10.235874112644488,
      "learning_rate": 4.925739315689991e-06,
      "loss": 1.0707824230194092,
      "memory(GiB)": 64.93,
      "step": 17,
      "token_acc": 0.7004186866460996,
      "train_speed(iter/s)": 0.006969
    },
    {
      "epoch": 9.0,
      "grad_norm": 10.235874112644488,
      "learning_rate": 4.903154239845798e-06,
      "loss": 1.063523530960083,
      "memory(GiB)": 64.93,
      "step": 18,
      "token_acc": 0.7153375032577535,
      "train_speed(iter/s)": 0.007202
    },
    {
      "epoch": 9.75294117647059,
      "grad_norm": 12.672585719832696,
      "learning_rate": 4.8776412907378845e-06,
      "loss": 1.0227261781692505,
      "memory(GiB)": 64.93,
      "step": 19,
      "token_acc": 0.7089719061524519,
      "train_speed(iter/s)": 0.006999
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.141126520279599,
      "learning_rate": 4.849231551964771e-06,
      "loss": 1.0062994956970215,
      "memory(GiB)": 64.93,
      "step": 20,
      "token_acc": 0.7253076155636847,
      "train_speed(iter/s)": 0.007198
    },
    {
      "epoch": 10.75294117647059,
      "grad_norm": 4.5706278016937,
      "learning_rate": 4.817959636416969e-06,
      "loss": 0.9886896014213562,
      "memory(GiB)": 64.93,
      "step": 21,
      "token_acc": 0.7129627749233363,
      "train_speed(iter/s)": 0.006977
    },
    {
      "epoch": 11.0,
      "grad_norm": 4.5706278016937,
      "learning_rate": 4.783863644106502e-06,
      "loss": 0.9500120878219604,
      "memory(GiB)": 64.93,
      "step": 22,
      "token_acc": 0.7324561403508771,
      "train_speed(iter/s)": 0.007164
    },
    {
      "epoch": 11.75294117647059,
      "grad_norm": 5.610030565636564,
      "learning_rate": 4.746985115747918e-06,
      "loss": 0.9478229284286499,
      "memory(GiB)": 64.93,
      "step": 23,
      "token_acc": 0.728414720452937,
      "train_speed(iter/s)": 0.00698
    },
    {
      "epoch": 12.0,
      "grad_norm": 5.610030565636564,
      "learning_rate": 4.707368982147318e-06,
      "loss": 0.9449450969696045,
      "memory(GiB)": 64.93,
      "step": 24,
      "token_acc": 0.733779116760967,
      "train_speed(iter/s)": 0.007149
    },
    {
      "epoch": 12.75294117647059,
      "grad_norm": 5.949700859685694,
      "learning_rate": 4.665063509461098e-06,
      "loss": 0.9186223149299622,
      "memory(GiB)": 64.93,
      "step": 25,
      "token_acc": 0.7384494558510042,
      "train_speed(iter/s)": 0.007
    },
    {
      "epoch": 13.0,
      "grad_norm": 2.7139067739877074,
      "learning_rate": 4.620120240391065e-06,
      "loss": 0.8890862464904785,
      "memory(GiB)": 64.93,
      "step": 26,
      "token_acc": 0.7381871086090201,
      "train_speed(iter/s)": 0.007153
    },
    {
      "epoch": 13.75294117647059,
      "grad_norm": 3.4053883533088602,
      "learning_rate": 4.572593931387604e-06,
      "loss": 0.8599684834480286,
      "memory(GiB)": 64.93,
      "step": 27,
      "token_acc": 0.7470102129905992,
      "train_speed(iter/s)": 0.007009
    },
    {
      "epoch": 14.0,
      "grad_norm": 3.4053883533088602,
      "learning_rate": 4.522542485937369e-06,
      "loss": 0.859759509563446,
      "memory(GiB)": 64.93,
      "step": 28,
      "token_acc": 0.7547843209648637,
      "train_speed(iter/s)": 0.007151
    },
    {
      "epoch": 14.75294117647059,
      "grad_norm": 3.616117900142953,
      "learning_rate": 4.470026884016805e-06,
      "loss": 0.8289841413497925,
      "memory(GiB)": 64.93,
      "step": 29,
      "token_acc": 0.7564268211182892,
      "train_speed(iter/s)": 0.00702
    },
    {
      "epoch": 15.0,
      "grad_norm": 3.616117900142953,
      "learning_rate": 4.415111107797445e-06,
      "loss": 0.824023962020874,
      "memory(GiB)": 64.93,
      "step": 30,
      "token_acc": 0.7594024374262875,
      "train_speed(iter/s)": 0.007149
    },
    {
      "epoch": 15.75294117647059,
      "grad_norm": 2.7367296058248014,
      "learning_rate": 4.357862063693486e-06,
      "loss": 0.8110586404800415,
      "memory(GiB)": 64.93,
      "step": 31,
      "token_acc": 0.7581217193928217,
      "train_speed(iter/s)": 0.007021
    },
    {
      "epoch": 16.0,
      "grad_norm": 2.153054302762085,
      "learning_rate": 4.2983495008466285e-06,
      "loss": 0.7662240266799927,
      "memory(GiB)": 64.93,
      "step": 32,
      "token_acc": 0.7881494613391518,
      "train_speed(iter/s)": 0.007149
    },
    {
      "epoch": 16.75294117647059,
      "grad_norm": 2.569194442983562,
      "learning_rate": 4.236645926147493e-06,
      "loss": 0.7760477066040039,
      "memory(GiB)": 64.93,
      "step": 33,
      "token_acc": 0.7681386710689229,
      "train_speed(iter/s)": 0.007021
    },
    {
      "epoch": 17.0,
      "grad_norm": 2.569194442983562,
      "learning_rate": 4.172826515897146e-06,
      "loss": 0.7777033448219299,
      "memory(GiB)": 64.93,
      "step": 34,
      "token_acc": 0.7749674539706156,
      "train_speed(iter/s)": 0.007139
    },
    {
      "epoch": 17.75294117647059,
      "grad_norm": 2.47234203743718,
      "learning_rate": 4.106969024216348e-06,
      "loss": 0.7580320239067078,
      "memory(GiB)": 64.93,
      "step": 35,
      "token_acc": 0.765316938732245,
      "train_speed(iter/s)": 0.007025
    },
    {
      "epoch": 18.0,
      "grad_norm": 2.47234203743718,
      "learning_rate": 4.039153688314146e-06,
      "loss": 0.7606703042984009,
      "memory(GiB)": 64.93,
      "step": 36,
      "token_acc": 0.77148792521423,
      "train_speed(iter/s)": 0.007137
    },
    {
      "epoch": 18.75294117647059,
      "grad_norm": 2.5562497660369816,
      "learning_rate": 3.969463130731183e-06,
      "loss": 0.7366118431091309,
      "memory(GiB)": 64.93,
      "step": 37,
      "token_acc": 0.7750171203561034,
      "train_speed(iter/s)": 0.00703
    },
    {
      "epoch": 19.0,
      "grad_norm": 1.4735735807130024,
      "learning_rate": 3.897982258676867e-06,
      "loss": 0.7414524555206299,
      "memory(GiB)": 64.93,
      "step": 38,
      "token_acc": 0.7816193202671952,
      "train_speed(iter/s)": 0.007136
    },
    {
      "epoch": 19.75294117647059,
      "grad_norm": 1.9640778001080272,
      "learning_rate": 3.824798160583012e-06,
      "loss": 0.719934344291687,
      "memory(GiB)": 64.93,
      "step": 39,
      "token_acc": 0.7814784727863525,
      "train_speed(iter/s)": 0.007035
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.9640778001080272,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.7130451202392578,
      "memory(GiB)": 64.93,
      "step": 40,
      "token_acc": 0.7878645026601118,
      "train_speed(iter/s)": 0.007139
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 28308993900544.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}